A RF-IPS Algorithm for Peer-to-Peer Video-on-Demand System

This paper presents an improved piece selection algorithm called RF-IPS algorithm, by analyzing the running mode of peer-to-peer (P2P) video-on-demand (VOD) systems and doing some research on a few piece selection algorithms which have been proposed in the past. The characteristic of our proposal is that we proposed "interval" based on the RF algorithm. Based on the chief feature of P2P VOD system, we partitioned the buffer into three intervals, the pieces in which we got through three different downloading strategies respectively. We carry through contrast experiments on piece selection algorithms by using five different strategies at the same conditions, and the results show our algorithm is efficient. Overall, we believe that the RF-IPS algorithm is feasible to further improve system performance and provide a cost-effective P2P VOD service with acceptable user experience.